Nemotron 3 Super-120B-A12B vs Qwen3-235B-A22B-Instruct-2507
Nemotron 3 Super-120B-A12B (2026) and Qwen3-235B-A22B-Instruct-2507 (2025) are frontier reasoning models from NVIDIA AI and Alibaba. Nemotron 3 Super-120B-A12B ships a 1M-token context window, while Qwen3-235B-A22B-Instruct-2507 ships a 256K-token context window. On LMArena Text Arena - Creative Writing, Qwen3-235B-A22B-Instruct-2507 leads by 58 pts. On pricing, Nemotron 3 Super-120B-A12B costs $0.09/1M input tokens versus $0.09/1M for the alternative. This comparison covers specs, pricing, API access, capabilities, benchmarks, input and output token costs, and production fit for coding and agent workloads.
Pick Qwen3-235B-A22B-Instruct-2507 for general evaluation; Nemotron 3 Super-120B-A12B is better when reasoning depth matters more.
Decision scorecard
Local evidence first| Signal | Nemotron 3 Super-120B-A12B | Qwen3-235B-A22B-Instruct-2507 |
|---|---|---|
| Best for | reasoning-heavy apps, tool-calling agents, and long-context analysis | provider-routed production |
| Decision fit | Coding, RAG, and Agents | RAG, Long context, and JSON / Tool use |
| Context window | 1M | 256K |
| Cheapest output | $0.40/1M tokens | $0.58/1M tokens |
| Provider routes | 6 tracked | 6 tracked |
| Shared benchmarks | 2 shared | LMArena Text Arena - Creative Writing leader |
Decision tradeoffs
- Nemotron 3 Super-120B-A12B holds a shared-benchmark lead on RULER, ahead by 2.9 points.
- Nemotron 3 Super-120B-A12B has the larger context window for long prompts, retrieval packs, or transcript analysis.
- Nemotron 3 Super-120B-A12B has the lower cheapest tracked output price at $0.40/1M tokens.
- Nemotron 3 Super-120B-A12B uniquely exposes Reasoning, JSON / Tool use, and Structured outputs in local model data.
- Local decision data tags Nemotron 3 Super-120B-A12B for Coding, RAG, and Agents.
- Qwen3-235B-A22B-Instruct-2507 holds a shared-benchmark lead on LMArena Text Arena - Creative Writing, ahead by 58 points.
- Local decision data tags Qwen3-235B-A22B-Instruct-2507 for RAG, Long context, and JSON / Tool use.
Monthly cost at traffic
Estimate token spend from the cheapest tracked input and output route or tier on this page.
Nemotron 3 Super-120B-A12B
$168
Cheapest tracked route/tier: DeepInfra
Qwen3-235B-A22B-Instruct-2507
$217
Cheapest tracked route/tier: Novita AI
Estimated monthly gap: $49.00. Batch, cache, alternate speed tiers, and negotiated pricing are excluded from this local estimate.
Switch friction
- Provider overlap exists on OpenRouter and Fireworks AI; start route-level A/B tests there.
- Qwen3-235B-A22B-Instruct-2507 is $0.18/1M tokens higher on cheapest tracked output pricing, so quality gains need to justify the spend.
- Check replacement coverage for Reasoning, JSON / Tool use, and Structured outputs before moving production traffic.
- Provider overlap exists on OpenRouter and Fireworks AI; start route-level A/B tests there.
- Nemotron 3 Super-120B-A12B is $0.18/1M tokens lower on cheapest tracked output pricing before cache, batch, or negotiated discounts.
- Nemotron 3 Super-120B-A12B adds Reasoning, JSON / Tool use, and Structured outputs in local capability data.
Specs
| Specification | ||
|---|---|---|
| Released | 2026-03-11 | 2025-07-21 |
| Context window | 1M | 256K |
| Parameters | 120B | 235B-A22B |
| Architecture | Decoder Only | Mixture of Experts |
| License | NVIDIA Nemotron Open Model License | Apache 2.0OSI-approved |
| Openness | Open weights | Open source |
| Weights | Available·NVIDIA Open Model | Available |
| Code | Unknown | Unknown |
| Commercial use | Permitted | Permitted |
| Knowledge cutoff | - | - |
Pricing and availability
| Pricing attribute | Nemotron 3 Super-120B-A12B | Qwen3-235B-A22B-Instruct-2507 |
|---|---|---|
| Input price | $0.09/1M tokens | $0.09/1M tokens |
| Output price | $0.40/1M tokens | $0.58/1M tokens |
| Providers |
Capabilities
| Capability | Nemotron 3 Super-120B-A12B | Qwen3-235B-A22B-Instruct-2507 |
|---|---|---|
| Vision | No | No |
| Multimodal | No | No |
| Reasoning | Yes | No |
| JSON / Tool use | Yes | No |
| Structured outputs | Yes | No |
| Code execution | No | No |
| IDE integration | No | No |
| Computer use | No | No |
| Parallel agents | No | No |
Benchmarks
| Benchmark | Nemotron 3 Super-120B-A12B | Qwen3-235B-A22B-Instruct-2507 |
|---|---|---|
| LMArena Text Arena - Creative Writing | 1317.0 | 1375.0 |
| RULER | 96.8 | 93.9 |
Deep dive
On shared benchmark coverage, LMArena Text Arena - Creative Writing has Nemotron 3 Super-120B-A12B at 1317 and Qwen3-235B-A22B-Instruct-2507 at 1375, with Qwen3-235B-A22B-Instruct-2507 ahead by 58 points; RULER has Nemotron 3 Super-120B-A12B at 96.8 and Qwen3-235B-A22B-Instruct-2507 at 93.9, with Nemotron 3 Super-120B-A12B ahead by 2.9 points. The largest visible gap is 58 points on LMArena Text Arena - Creative Writing, which matters most when that benchmark mirrors your workload. Treat isolated benchmark wins as directional, because provider routing, prompt style, and tool access can move real application results.
The capability footprint differs most on reasoning mode: Nemotron 3 Super-120B-A12B, JSON / Tool use: Nemotron 3 Super-120B-A12B, and structured outputs: Nemotron 3 Super-120B-A12B. Both models share the core language-model surface, so the practical split is not just feature count. Use those differences to decide whether the page is about raw model quality, agentic coding support, multimodal ingestion, or predictable structured API behavior.
For cost, Nemotron 3 Super-120B-A12B lists $0.09/1M input and $0.40/1M output tokens on the cheapest listed provider, while Qwen3-235B-A22B-Instruct-2507 lists $0.09/1M input and $0.58/1M output tokens on the cheapest listed provider. A 70/30 input-output blend puts Nemotron 3 Super-120B-A12B lower by about $0.06 per million blended tokens. Availability is 6 providers versus 6, so concentration risk also matters.
Choose Nemotron 3 Super-120B-A12B when reasoning depth, larger context windows, and lower input-token cost are central to the workload. Choose Qwen3-235B-A22B-Instruct-2507 when provider fit are more important. For production, rerun your own prompts through the exact provider, region, and tool stack you plan to ship.
FAQ
Which has a larger context window, Nemotron 3 Super-120B-A12B or Qwen3-235B-A22B-Instruct-2507?
Nemotron 3 Super-120B-A12B supports 1M tokens, while Qwen3-235B-A22B-Instruct-2507 supports 256K tokens. That gap matters most for long documents, large codebases, retrieval-heavy agents, and conversations where earlier context must remain visible.
Which is cheaper, Nemotron 3 Super-120B-A12B or Qwen3-235B-A22B-Instruct-2507?
Nemotron 3 Super-120B-A12B is cheaper on listed token pricing. Nemotron 3 Super-120B-A12B costs $0.09/1M input and $0.40/1M output tokens. Qwen3-235B-A22B-Instruct-2507 costs $0.09/1M input and $0.58/1M output tokens. Provider discounts or batch pricing can still change the final bill.
Is Nemotron 3 Super-120B-A12B or Qwen3-235B-A22B-Instruct-2507 open source?
Nemotron 3 Super-120B-A12B is listed under NVIDIA Nemotron Open Model License. Qwen3-235B-A22B-Instruct-2507 is listed under Apache 2.0. License labels affect whether you can self-host, redistribute weights, or rely only on hosted APIs, so confirm the upstream license before deployment.
Which is better for reasoning mode, Nemotron 3 Super-120B-A12B or Qwen3-235B-A22B-Instruct-2507?
Nemotron 3 Super-120B-A12B has the clearer documented reasoning mode signal in this comparison. If reasoning mode is mission-critical, validate it against the provider endpoint because model-level support and API-level exposure can differ.
Which is better for JSON / Tool use, Nemotron 3 Super-120B-A12B or Qwen3-235B-A22B-Instruct-2507?
Nemotron 3 Super-120B-A12B has the clearer documented JSON / Tool use signal in this comparison. If JSON / Tool use is mission-critical, validate it against the provider endpoint because model-level support and API-level exposure can differ.
Where can I run Nemotron 3 Super-120B-A12B and Qwen3-235B-A22B-Instruct-2507?
Nemotron 3 Super-120B-A12B is available on Cloudflare Workers AI, DeepInfra, NVIDIA NIM, OpenRouter, and Fireworks AI. Qwen3-235B-A22B-Instruct-2507 is available on AWS Bedrock, Novita AI, OpenRouter, GCP Vertex AI, and Together AI. Provider coverage can affect latency, region availability, compliance posture, and fallback options.
Continue comparing
Last reviewed: 2026-10-26. Data sourced from public model cards and provider documentation.